and Laboratory Methods. 1375 



agent, on the other hand, gives rise to skeleton crystals and to masses of 

 thin lenticular plates. 



During the disintegration of the fragment of potassium oxalate while passing 

 into solution (particularly in concentrated solutions), crystals of the reagent 

 appear mof/w/itari/y, which bear a striking resemblance to some of the forms 

 assumed by the double glucinum potassium oxalate. In testing unknown solu- 

 tions the worker must be on his guard lest he fall into error by deciding too 

 hastily. 



The double oxalate of glucinum and potassium can be readily recrystallized from 

 water by gently warming the preparation and allowing it to cool slowly. The 

 salt is also soluble in solutions of ammonium carbonate, a property which can 

 be utilized when there is doubt as to the nature of the precipitate obtained in the 

 course of an analysis. 



The addition of a little mercuric chloride will induce the production of long 

 prisms and twins, and hence is useful when good crystals cannot otherwise be 

 obtained. 



Neither primary potassium oxalate, sodium, nor ammonium oxalates can be 

 substituted for the normal oxalate of potassium. 



With zinc, the reagent gives tiny double globulites and pseudo-octahedra of 

 normal zinc oxalate, and later, as the test drop concentrates by evaporation, 

 neat hexagonal plates appear, which are probably due to a double oxalate of 

 potassium and zinc (?). Mercuric chloride seems to favor the formation of the 

 hexagonal plates. 



Cadmium treated in like manner yields, apparently, only crystals of normal 

 cadmium oxalate (g. v.). No double salt seems to separate. 



The reagent gives nothing with magnesium, providing the test drop is not too 

 concentrated and does not contain an excessive amount of free acetic acid. 



When zinc or cadmium is also present, the crystal form of the glucinum 

 potassium oxalate is changed. It then becomes difficult to decide whether or 

 not glucinum is present. 



Magnesium, aluminum, and iron, on the other hand, have practically no 

 influence, unless present in relatively large amount. But the double oxalate of 

 glucinum and potassium crystallizing from such solutions will always occlude an 

 appreciable quantity of the potassium double oxalate of these elements. 



Calcium, strontium, and barium may mask the reaction. 



Ammonium salts, if present, must first be removed by gentle ignition before 

 testing with potassium oxalate. 



Free mineral acids must be absent. 



Stannous salts may at times give, with potassium oxalate, crystals of stannous 

 oxalate which may be mistaken by an inexperienced worker for the glucinum 

 double salt.* After the tin salt has been allowed to grow for a short time, 

 there is little danger of confusing the two. If still in doubt, recrystallize from 

 warm water, treat with ammonium carbonate, or apply tests for tin. 



When the solution to be tested contains copper, cobalt, or nickel, it is gener- 



* This error cannot arise in the course of a systematic analysis. 



